Commit Graph

370 Commits

Author SHA1 Message Date
Chris Lattner
ddf89566a9 This commit changes:
1. Legalize now always promotes truncstore of i1 to i8. 
2. Remove patterns and gunk related to truncstore i1 from targets.
3. Rename the StoreXAction stuff to TruncStoreAction in TLI.
4. Make the TLI TruncStoreAction table a 2d table to handle from/to conversions.
5. Mark a wide variety of invalid truncstores as such in various targets, e.g.
   X86 currently doesn't support truncstore of any of its integer types.
6. Add legalize support for truncstores with invalid value input types.
7. Add a dag combine transform to turn store(truncate) into truncstore when
   safe.

The later allows us to compile CodeGen/X86/storetrunc-fp.ll to:

_foo:
	fldt	20(%esp)
	fldt	4(%esp)
	faddp	%st(1)
	movl	36(%esp), %eax
	fstps	(%eax)
	ret

instead of:

_foo:
	subl	$4, %esp
	fldt	24(%esp)
	fldt	8(%esp)
	faddp	%st(1)
	fstps	(%esp)
	movl	40(%esp), %eax
	movss	(%esp), %xmm0
	movss	%xmm0, (%eax)
	addl	$4, %esp
	ret



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46140 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-17 19:59:44 +00:00
Chris Lattner
4626b250a2 code cleanups, no functionality change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46126 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-17 07:20:38 +00:00
Chris Lattner
0bd4893a07 * Introduce a new SelectionDAG::getIntPtrConstant method
and switch various codegen pieces and the X86 backend over
  to using it.

* Add some comments to SelectionDAGNodes.h

* Introduce a second argument to FP_ROUND, which indicates
  whether the FP_ROUND changes the value of its input. If
  not it is safe to xform things like fp_extend(fp_round(x)) -> x.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46125 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-17 07:00:52 +00:00
Evan Cheng
02c4285643 Fixes a nasty dag combiner bug that causes a bunch of tests to fail at -O0.
It's not safe to use the two value CombineTo variant to combine away a dead load.
e.g. 
v1, chain2 = load chain1, loc
v2, chain3 = load chain2, loc
v3         = add v2, c 
Now we replace use of v1 with undef, use of chain2 with chain1.
ReplaceAllUsesWith() will iterate through uses of the first load and update operands:
v1, chain2 = load chain1, loc
v2, chain3 = load chain1, loc
v3         = add v2, c 
Now the second load is the same as the first load, SelectionDAG cse will ensure
the use of second load is replaced with the first load.
v1, chain2 = load chain1, loc
v3         = add v1, c
Then v1 is replaced with undef and bad things happen.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46099 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-16 23:11:54 +00:00
Chris Lattner
572dee71af Factor the ReachesChainWithoutSideEffects out of dag combiner into
a public SDOperand::reachesChainWithoutSideEffects method.  No 
functionality change.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46050 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-16 05:49:24 +00:00
Chris Lattner
07649d9265 Make load->store deletion a bit smarter. This allows us to compile this:
void test(long long *P) { *P ^= 1; }

into just:

_test:
	movl	4(%esp), %eax
	xorl	$1, (%eax)
	ret

instead of code like this:

_test:
	movl	4(%esp), %ecx
        xorl    $1, (%ecx)
	movl	4(%ecx), %edx
	movl	%edx, 4(%ecx)
	ret



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45762 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-08 23:08:06 +00:00
Chris Lattner
4ee451de36 Remove attribution from file headers, per discussion on llvmdev.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45418 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29 20:36:04 +00:00
Chris Lattner
b253a8b382 make sure not to zap volatile stores, thanks a lot to Dale for noticing this!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45402 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29 07:15:45 +00:00
Chris Lattner
5938bef2d2 don't fold fp_round(fp_extend(load)) -> fp_round(extload)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45400 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29 06:55:23 +00:00
Chris Lattner
3c87285af6 Delete a store whose input is a load from the same pointer:
x = load p
  store x -> p


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45398 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29 06:26:16 +00:00
Chris Lattner
b16f55f3c8 Tell TargetLoweringOpt whether it is running before
or after legalize.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45321 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-22 20:56:36 +00:00
Evan Cheng
02132d62c1 Don't leave newly created nodes around if it turns out they are not needed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45186 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-19 01:34:38 +00:00
Dale Johannesen
e0e6fac5e9 Redo previous patch so optimization only done for i1.
Simpler and safer.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44663 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 17:53:31 +00:00
Chris Lattner
d3fd6d2b25 third time around: instead of disabling this completely,
only disable it if we don't know it will be obviously profitable.
Still fixme, but less so. :)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44658 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 07:47:55 +00:00
Chris Lattner
e00cbad78d Actually, disable this code for now. More analysis and improvements to
the X86 backend are needed before this should be enabled by default.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44657 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 07:44:31 +00:00
Chris Lattner
e70da20f9a implement a readme entry, compiling the code into:
_foo:
	movl	$12, %eax
	andl	4(%esp), %eax
	movl	_array(%eax), %eax
	ret

instead of:

_foo:
	movl	4(%esp), %eax
	shrl	$2, %eax
	andl	$3, %eax
	movl	_array(,%eax,4), %eax
	ret

As it turns out, this triggers all the time, in a wide variety of
situations, for example, I see diffs like this in various programs:

-       movl    8(%eax), %eax
-       shll    $2, %eax
-       andl    $1020, %eax
-       movl    (%esi,%eax), %eax
+       movzbl  8(%eax), %eax
+       movl    (%esi,%eax,4), %eax


-       shll    $2, %edx
-       andl    $1020, %edx
-       movl    (%edi,%edx), %edx
+       andl    $255, %edx
+       movl    (%edi,%edx,4), %edx

Unfortunately, I also see stuff like this, which can be fixed in the
X86 backend:

-       andl    $85, %ebx
-       addl    _bit_count(,%ebx,4), %ebp
+       shll    $2, %ebx
+       andl    $340, %ebx
+       addl    _bit_count(%ebx), %ebp



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44656 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 07:33:36 +00:00
Dale Johannesen
59f1e97ef3 Fix PR1842.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44649 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-06 01:43:46 +00:00
Dan Gohman
77003040b2 Don't lower srem/urem X%C to X-X/C*C unless the division is actually
optimized. This avoids creating illegal divisions when the combiner is
running after legalize; this fixes PR1815. Also, it produces better
code in the included testcase by avoiding the subtract and multiply
when the division isn't optimized.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44341 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-26 23:46:11 +00:00
Duncan Sands
fd617d0143 Move MinAlign to MathExtras.h.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43944 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-09 13:41:39 +00:00
Duncan Sands
c6fa170b4d Fix some load/store logic that would be wrong for
apints on big-endian machines if the bitwidth is
not a multiple of 8.  Introduce a new helper,
MVT::getStoreSizeInBits, and use it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43934 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-09 08:57:19 +00:00
Evan Cheng
4471194d69 If both parts of smul_lohi, etc. are used, don't simplify. If only one part is used, try simplify it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43888 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-08 09:25:29 +00:00
Evan Cheng
de1631bc85 Typo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43511 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-30 20:11:21 +00:00
Dan Gohman
090b38a0da Fix a DAGCombiner abort on a bitcast from a scalar to a vector.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43470 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-29 20:44:42 +00:00
Evan Cheng
3c3ddb3a85 Enable more fold (sext (load x)) -> (sext (truncate (sextload x)))
transformation. Previously, it's restricted by ensuring the number of load uses
is one. Now the restriction is loosened up by allowing setcc uses to be
"extended" (e.g. setcc x, c, eq -> setcc sext(x), sext(c), eq).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43465 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-29 19:58:20 +00:00
Duncan Sands
dc84650679 The guaranteed alignment of ptr+offset is only the minimum of
of offset and the alignment of ptr if these are both powers of
2.  While the ptr alignment is guaranteed to be a power of 2,
there is no reason to think that offset is.  For example, if
offset is 12 (the size of a long double on x86-32 linux) and
the alignment of ptr is 8, then the alignment of ptr+offset
will in general be 4, not 8.  Introduce a function MinAlign,
lifted from gcc, for computing the minimum guaranteed alignment.
I've tried to fix up everywhere under lib/CodeGen/SelectionDAG/.
I also changed some places that weren't wrong (because both values
were a power of 2), as a defensive change against people copying
and pasting the code.
Hopefully someone who cares about alignment will review the rest
of LLVM and fix up the remaining places.  Since I'm on x86 I'm
not very motivated to do this myself...


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43421 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-28 12:59:45 +00:00
Dale Johannesen
b6210fc92b Redo "last ppc long double fix" as Chris wants.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43189 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-19 20:29:00 +00:00
Dale Johannesen
fabd32deb0 More ppcf128 issues (maybe the last)?
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43160 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-19 00:59:18 +00:00
Dale Johannesen
db44bf85d8 Disable attempts to constant fold PPC f128.
Remove the assumption that this will happen from
various places.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43053 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-16 23:38:29 +00:00
Chris Lattner
01d029b82c One mundane change: Change ReplaceAllUsesOfValueWith to *optionally*
take a deleted nodes vector, instead of requiring it.

One more significant change:  Implement the start of a legalizer that
just works on types.  This legalizer is designed to run before the 
operation legalizer and ensure just that the input dag is transformed
into an output dag whose operand and result types are all legal, even
if the operations on those types are not.

This design/impl has the following advantages:

1. When finished, this will *significantly* reduce the amount of code in
   LegalizeDAG.cpp.  It will remove all the code related to promotion and
   expansion as well as splitting and scalarizing vectors.
2. The new code is very simple, idiomatic, and modular: unlike 
   LegalizeDAG.cpp, it has no 3000 line long functions. :)
3. The implementation is completely iterative instead of recursive, good
   for hacking on large dags without blowing out your stack.
4. The implementation updates nodes in place when possible instead of 
   deallocating and reallocating the entire graph that points to some 
   mutated node.
5. The code nicely separates out handling of operations with invalid 
   results from operations with invalid operands, making some cases
   simpler and easier to understand.
6. The new -debug-only=legalize-types option is very very handy :), 
   allowing you to easily understand what legalize types is doing.

This is not yet done.  Until the ifdef added to SelectionDAGISel.cpp is
enabled, this does nothing.  However, this code is sufficient to legalize
all of the code in 186.crafty, olden and freebench on an x86 machine.  The
biggest issues are:

1. Vectors aren't implemented at all yet
2. SoftFP is a mess, I need to talk to Evan about it.
3. No lowering to libcalls is implemented yet.
4. Various operations are missing etc.
5. There are FIXME's for stuff I hax0r'd out, like softfp.

Hey, at least it is a step in the right direction :).  If you'd like to help,
just enable the #ifdef in SelectionDAGISel.cpp and compile code with it.  If
this explodes it will tell you what needs to be implemented.  Help is 
certainly appreciated.

Once this goes in, we can do three things:

1. Add a new pass of dag combine between the "type legalizer" and "operation
   legalizer" passes.  This will let us catch some long-standing isel issues
   that we miss because operation legalization often obfuscates the dag with
   target-specific nodes.
2. We can rip out all of the type legalization code from LegalizeDAG.cpp,
   making it much smaller and simpler.  When that happens we can then 
   reimplement the core functionality left in it in a much more efficient and
   non-recursive way.
3. Once the whole legalizer is non-recursive, we can implement whole-function
   selectiondags maybe...



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42981 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-15 06:10:22 +00:00
Chris Lattner
e33544ce55 Enhance the truncstore optimization code to handle shifted
values and propagate demanded bits through them in simple cases.

This allows this code:
void foo(char *P) {
   strcpy(P, "abc");
}
to compile to:

_foo:
        ldrb r3, [r1]
        ldrb r2, [r1, #+1]
        ldrb r12, [r1, #+2]!
        ldrb r1, [r1, #+1]
        strb r1, [r0, #+3]
        strb r2, [r0, #+1]
        strb r12, [r0, #+2]
        strb r3, [r0]
        bx lr

instead of:

_foo:
        ldrb r3, [r1, #+3]
        ldrb r2, [r1, #+2]
        orr r3, r2, r3, lsl #8
        ldrb r2, [r1, #+1]
        ldrb r1, [r1]
        orr r2, r1, r2, lsl #8
        orr r3, r2, r3, lsl #16
        strb r3, [r0]
        mov r2, r3, lsr #24
        strb r2, [r0, #+3]
        mov r2, r3, lsr #16
        strb r2, [r0, #+2]
        mov r3, r3, lsr #8
        strb r3, [r0, #+1]
        bx lr

testcase here: test/CodeGen/ARM/truncstore-dag-combine.ll

This also helps occasionally for X86 and other cases not involving 
unaligned load/stores.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42954 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-13 06:58:48 +00:00
Chris Lattner
2b4c279a8e Add a simple optimization to simplify the input to
truncate and truncstore instructions, based on the 
knowledge that they don't demand the top bits.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42952 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-13 06:35:54 +00:00
Duncan Sands
272dce0137 Correct swapped arguments to getConstant.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42824 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-10 09:54:50 +00:00
Dan Gohman
389079b59f DAGCombiner support for UDIVREM/SDIVREM and UMUL_LOHI/SMUL_LOHI.
Check if one of the two results unneeded so see if a simpler operator
could bs used. Also check to see if each of the two computations could be
simplified if they were split into separate operators. Factor out the code
that calls visit() so that it can be used for this purpose.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42759 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-08 17:57:15 +00:00
Evan Cheng
513da43c9c Reapply 42677.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42692 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-06 08:19:55 +00:00
Chris Lattner
5349de1372 revert evan's patch until the header is committed
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42686 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-06 06:08:17 +00:00
Evan Cheng
c231e8c8a5 Added DAG xforms. e.g.
(vextract (v4f32 s2v (f32 load $addr)), 0) -> (f32 load $addr) 
(vextract (v4i32 bc (v4f32 s2v (f32 load $addr))), 0) -> (i32 load $addr)
Remove x86 specific patterns.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42677 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-06 02:46:29 +00:00
Evan Cheng
e0480d2ec2 Fix a bogus splat xform:
shuffle <undef, undef, x, undef>, <undef, undef, undef, undef>, <2, 2, 2, 2>
!=
<undef, undef, x, undef>


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42111 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-18 21:54:37 +00:00
Dale Johannesen
c7b21d520a Prevent crash on long double.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42103 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-18 18:36:59 +00:00
Dale Johannesen
9d5f456077 Revise previous patch per review comments.
Next round of x87 long double stuff.
Getting close now, basically works.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41875 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-12 03:30:33 +00:00
Dale Johannesen
3f6eb7419d Add APInt interfaces to APFloat (allows directly
access to bits).  Use them in place of float and
double interfaces where appropriate.
First bits of x86 long double constants handling 
(untested, probably does not work).



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41858 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-11 18:32:33 +00:00
Chris Lattner
61c5ff460b Emit:
cmpl    %eax, %ecx
        setae   %al
        movzbl  %al, %eax

instead of:

        cmpl    %eax, %ecx
        setb    %al
        xorb    $1, %al
        movzbl  %al, %eax

when using logical not of a C comparison.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41807 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-10 21:39:07 +00:00
Dale Johannesen
c4dd3c3b51 Add mod, copysign, abs operations to APFloat.
Implement some constant folding in SelectionDAG and
DAGCombiner using APFloat.  Remove double versions
of constructor and getValue from ConstantFPSDNode.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41664 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-31 23:34:27 +00:00
Dan Gohman
e9c8fa095e Make DAGCombiner's global alias analysis query more precise in the case
where both pointers have non-zero offsets.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41491 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27 16:32:11 +00:00
Dale Johannesen
e6c1742914 Revise per review comments.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41409 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-26 01:18:27 +00:00
Dale Johannesen
87503a63d5 Add APFloat interface to ConstantFPSDNode. Change
over uses in DAGCombiner.  Fix interfaces to work
with APFloats.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41407 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-25 22:10:57 +00:00
Evan Cheng
571c4788d5 Fold C ? 0 : 1 to ~C or zext(~C) or trunc(~C) depending the types.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41163 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-18 05:57:05 +00:00
Dan Gohman
559742c0ea Fix the alias analysis query in DAGCombiner to not add in two
offsets. The SrcValueOffset values are the real offsets from the
SrcValue base pointers.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40534 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-26 16:14:06 +00:00
Dan Gohman
05d92fe7c4 Don't call SimplifyVBinOp for non-vector operations, following earlier review
feedback. This theoretically makes the common (scalar) case more efficient.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@39823 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-13 20:03:40 +00:00
Dan Gohman
70fb1aefd5 Fix a bug in the folding of binary operators to undef.
Thanks to Lauro for spotting this!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@38491 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-10 15:19:29 +00:00
Dan Gohman
d595b5f1f0 Fix the folding of undef in several binary operators to recognize
undef in either the left or right operand.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@38489 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-10 14:20:37 +00:00